Analyst View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
PAPER
Cite this: DOI: 10.1039/c5an00826c
View Journal
Effects of charge states, charge sites and side chain interactions on conformational preferences of a series of model peptide ions† Chunying Xiao, Lisa M. Pérez and David H. Russell* The effects of charge states, charge sites and side chain interactions on conformational preferences of gas-phase peptide ions are examined by ion mobility-mass spectrometry (IM-MS) and molecular dynamics (MD) simulations. Collision cross sections (CCS) of [M + 2H]2+ and [M + 3H]3+ ions for a series of model peptides, viz. Ac-(AAKAA)nY-NH2 (AKn, n = 3–5) and Ac-Y(AEAAKA)nF-NH2 (AEKn, n = 2–5) are measured by using IM-MS and compared with calculated CCS for candidate ions generated by MD simu-
Received 27th April 2015, Accepted 10th June 2015 DOI: 10.1039/c5an00826c www.rsc.org/analyst
lations. The results show that charge states, charge sites and intramolecular charge solvation are important determinants of conformer preference for AKn and AEKn ions. For AKn ions, there is a strong preference for helical conformations near the N-terminus and charge-solvated conformations near the C-terminus. For [AEKn + 2H]2+ ions, conformer preferences appear to be driven by charge solvation, whereas [AEKn + 3H]3+ ions favor more extended coil-type conformations.
Introduction Peptides and proteins are highly dynamic, sampling many conformations on rapid time scales, and both the dynamics and conformational preferences are sensitive to the local environment.1–4 Conformational preferences and the dynamics of interconversions among the different conformations are highly dependent on intramolecular interactions, i.e., van der Waals interactions, hydrogen bonding, and electrostatic interactions. The presence of water as well as cations and anions also plays a key role in determining conformational preferences, both in terms of the hydrophobic effects as well as hydrophilic interactions, viz. solvent-accessibility to hydrophilic side chains of histidine, lysine, arginine, and interactions (salt-bridges) of these charge sites with oppositely charged aspartate and glutamate (R–COO−) as well as asparagine and glutamine.5,6 The effects of intramolecular hydrophilic interactions on the conformational preferences of peptides/proteins have been extensively studied7–9 and advances in experimental techniques and molecular dynamics (MD) simulations are providing new approaches to studies of both thermodynamic stability and folding/unfolding kinetics of native and non-native conformations.8,10 Although MD simulations are widely used for studies of peptide/protein con-
Texas A&M University, Department of Chemistry, College Station, Texas 77843, USA. E-mail: [email protected] † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c5an00826c
This journal is © The Royal Society of Chemistry 2015
formational preferences, simulations only provide candidate conformations that must be evaluated against experimental data, i.e., from circular dichroism (CD),11 fluorescence and fluorescence resonance energy transfer (FRET),12,13 nuclear magnetic resonance (NMR),14,15 X-ray diffraction (XRD),14 and small-angle X-ray scattering (SAXS).16–18 There is increasing awareness that mass spectrometry (MS) approaches add new dimensions for understanding peptide/protein structure and extending our understanding of structure/function relationships, especially for studies of systems that are composed of multiple conformations. MS-based approaches for studies of peptide/protein conformations have evolved to include H/D exchange,19,20 chemical labeling,21,22 IR-UV double resonance spectroscopy23,24 and tandem MS.25,26 IM-MS combined with MD simulation complements the more traditional approaches mentioned above27 since it provides a direct determination of the conformational heterogeneity of the ion population, and it can be adapted to high throughput workflows for screening complex peptide/ protein libraries.25,28 Electrospray ionization (ESI) is clearly the ionization method of choice for structural MS since it has been shown to maintain noncovalent interactions,29–32 and even to retain solution phase secondary structure.33 Recently, Breuker and coworkers studied protein structures in the solution and gas phase, and demonstrated that salt bridges can help to stabilize native/compact structure on short time scales when the proteins are transferred from solution phase to the gas phase.34,35 A significant advantage of ESI is the ability to produce mass spectra dominated by multiply-charged ions;
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
however, it is difficult to determine the location of the charge for molecules that contain multiple possible charge sites. IM-MS is based on the measurement of ion-neutral CCS, which reflects the size and three-dimensional shape of an ion, and provides a direct measurement on population heterogeneity, viz. the number of peaks in the IM arrival-time distributions (ATD). A priori assignment of IM CCSs is a daunting task, viz., currently there exist no guiding principles for correlating the measured CCS for an ion to specific 3-D shapes. In some cases it is possible to correlate CCS with 3-D shapes derived from other experimental measurements, specifically XRD or NMR;36–39 however, MD simulations provide a broader sampling of candidate conformations of the ions whose CCS values can be calculated by using MOBCAL40 (or similar methods).36,41 The “theoretical” CCS obtained using MD simulations and MOBCAL can then be compared to the experimental CCS. This study employs two series of peptides to illustrate the data mining made possible by integration of IM-MS and MD simulation. The peptide sequences are Ac-(AAKAA)nY-NH2 (AKn series, n = 3–5) and Ac-Y(AEAAKA)nF-NH2 (AEKn series, n = 2–5); Ac- indicates acetylation at the N-termini and –NH2 indicates amidation at the C-termini. The N-termini and C-termini of the peptides in each series were protected to reduce the likelihood of collapse of the helix by having charge(s) located at the termini.42 The model peptides used in this study have been used previously by several groups to probe helical propensities of alanine-containing peptides in solution43,44 and in low dielectric gas-phase (solvent-free) environments.45 Previous CD studies show that in solution, the helical content of both AKn and AEKn increases as n increases,43,44,46 and the helical content of AEKn is higher than that of AKn for peptides of comparable length owing to the stabilization afforded by salt bridges. In prior IM-MS studies of the singly-charged ions for the two peptides, it was found that gas-phase ions of AKn (n = 3–6) have higher helical content (ca. 60% helical) than AEKn, and helical propensity is the highest when charge is aligned with the helix macrodipole, i.e., the charge is located on or near the C-termini.45 Despite extensive studies of the AKn and AEKn series, the following questions still remain unanswered: (i) how do charge sites and charge states affect conformer preferences; (ii) how do intramolecular interactions involving the polar side chains affect conformer preferences; and (iii) is there any relationship between solution-phase and gas-phase conformations? Here, ESI-IM-MS in conjunction with MD simulations was used to evaluate the conformational preferences of the two model peptide series in an effort to address these questions.
Experimental section Sample preparation Model peptides AKn (n = 3–5) and AEKn (n = 2–5) were purchased from Shanghai Mocell Biotech Co., Ltd and used without further purification. Solutions of the model peptides
Analyst
Analyst
(2 μM) were prepared using 10 : 90 (v/v) methanol/buffer (buffer: 50 mM ammonium acetate, 25 mM imidazole; pH was adjusted to 7 with 0.1 M acetic acid). ESI-IM-MS All mobility profiles were acquired on a Waters Synapt HDMS G2 mass spectrometer (Manchester, U.K.). Ions were produced by nano-ESI with a source temperature of 120 °C and capillary voltage of 1.5 kV, sampling cone voltage of 40 V and extraction cone voltage of 4 V. For ion mobility experiments, the traveling wave ion mobility cell was maintained at 3.01 mbar N2. The travelling wave velocity is 500 m s−1 and wave height is 20 V for [AKn + 3H]3+; wave velocity and wave height are 300 m s−1 and 20 V respectively for both [AKn + 2H]2+ and [AEKn + 3H]3+; wave velocity and wave height are 800 m s−1 and 40 V respectively for [AEKn + 2H]2+ due to the difference in mobility of different peptides. CCS calibration was performed as previously described by Ruotolo et al.47 Tryptically digested peptides of cytochrome c and myoglobin were used as calibration standards and literature CCS values of doubly-charged peptide ions are cited from the Clemmer group’s CCS database.48 It should be noted that the triply-protonated model peptides are also calibrated using doubly-charged ions because currently there is no standard calibrant database for triply-protonated peptide ions. To probe the stability of the model peptides, solvents conditions were screened to ensure CCS profiles are not influenced by the solvents (data not shown). Instrumental conditions such as capillary voltage, sampling cone voltage and extraction cone voltage, were also screened to ensure arrival time distribution (ATD) did not change under various conditions. Calculation of helical content and number of helical residues Helical content ð% Þ ¼
Ωobs Ωglob Ωhelix Ωglob
100
ð1Þ
Number of helical residues ¼ Helical content total residue number
ð2Þ
Ωobs is the CCS determined by the centroid of the peak profiles. Ωhelix is the calculated CCS for a rigid α-helix cited from literature.45 Ωglob is the calculated globular CCS using the standard globular trend line equations: y ¼ 0:1364x þ 105:99 y ¼ 0:1272x þ 134
For þ 2 species For þ 3 species
ð3Þ ð4Þ
here, x is the mass of peptides, and y is the CCS of globular structure. The standard trend line for doubly-protonated peptide ions (grey in Fig. 1B and 3B, upper) was drawn based on the literature data of doubly-charged ions.48 The standard trend line for triply-protonated ions (grey in Fig. 1B and 3B, lower) was drawn based on standard triply-protonated peptide ions pre-
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Analyst
Paper
Fig. 1 CCS profiles and helical content for AKn. A. CCS profiles for AKn ions. Green peaks are the deconvoluted peak profiles for [AK3 + 2H]2+. “A” and “B” are the higher and lower abundance peaks for [AK3 + 2H]2+. B. CCS vs. molecular weight of AKn ions. Grey points are standard peptide data representing CCS for peptide ions that prefer compact globular conformations in the gas-phase. The doubly-charged standard data is cited from the database of the Clemmer group and triply-charged standard data are from digested proteins. C. Helical content of AKn ions in the solution phase (SP, black) and gas-phase (GP, blue).
pared from the miscleavage of trypsin digestion of proteins (ubiquitin, lysozyme (from chicken egg white), aldolase, enolase, avidin (from egg white), albumin (from chicken egg), ribonuclease (bovin pancrease), myoglobin (horse), cytocrome c, BSA (horse)). All proteins were obtained from Sigma-Aldrich and used without further purification. Molecular dynamics (MD) simulations Candidate conformations of AKn and AEKn ions were generated by simulated annealing (SA) using AMBER1149 with the FF99SB force field.50 For the AKn and AEKn peptide series, both the N-termini and C-termini are protected (acetylated and amidated, respectively), thus the lysine side chains, which have high proton affinities, were considered as the charge carrying sites. In cases where the number of lysines exceeded the numbers of protons, all possible proton positions were considered. The α-helical and fully extended conformations were used as starting structures for SA. During one annealing cycle, the system was heated from 300 K to 1000 K and then cooled down to 300 K over 8400 fs with a time step of 1 fs. After each annealing cycle, the structure energy was minimized. SA was performed for 300 annealing cycles producing 300 minimized structures per simulation. To ensure proper sampling of the
This journal is © The Royal Society of Chemistry 2015
conformational space, a second set of simulations was performed starting with the lowest energy structures from the first set of simulations. For [AEK4 + 2H]2+, two sets of simulations did not produce enough sampling, i.e., the candidate structures did not vary much, therefore, a third set of simulations was performed. The highest energy conformation and a randomly selected middle energy conformation with CCSs in the range of ±2% of the experimental value from the first and second set of simulations were chosen as the starting structures of the third set of simulations. CCSs were calculated by the trajectory method in MOBCAL.40 The secondary structure was calculated using DSSP program.51,52 When processing the simulation data, filtering by CCS and RMSD cluster analysis was performed on the candidate conformations. The procedure for making clusters is described in Fig. S1 (see ESI†). First, all generated conformations that have CCSs within ±2% of the experimental CCS value for the doubly-charged ions form family 1 (in the case of triplycharged ions, ±5% of the experimental data form family 1 because the triply-charged ions were calibrated by doublycharged standard ions which may result in higher error). Structures in family 1 were then clustered based on backbone RMSD (Root-mean-square deviation). Cluster 1 consists of all
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
conformations in family 1 that have an RMSD of the backbone atoms less than a specified value (RMSD cutoff is typically 4–4.5 Å2) when compared to the lowest energy structure in family 1. All of the conformations with an RMSD higher than the specified value make family 2. Cluster 2 contains all structures with an RMSD of the backbone atoms below the cutoff when compared to the lowest energy structure in family 2 and all the conformations with an RMSD higher than the cutoff make family 3. The procedure was repeated for the remaining structures until all the structures were allocated into a cluster. Therefore, the reference structure for cluster 1 has the lowest energy, the reference structure for cluster 2 has higher energy than that for cluster 1 and the reference structure for the last cluster has the highest energy. It should be noted that due to the different size of peptides, the RMSD cutoff was adjusted to produce a reasonable number of clusters ( m/2 denotes that the net charge is located on the C-terminal half (C)
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Theoretical net charge distribution (TNCD) CS
+2
Peptides N (%) M (%) C (%)
+3 AK3
AK4
AK5
AEK3
AEK4
AEK3EC
AEK4EC
AK4
AK5
AEK4
AEK4EC
33 33 33
33 33 33
40 20 40
33 — 67
33 — 67
— — 100
17 — 83
50 — 50
40 20 40
25 0 75
0 25 75
AEK4EC
AK4
AK5
AEK4
AEK4EC
57 — 43
46 20 34
25 — 75
— 25 75
Group A (NCD) CS
+2
+3
Peptides
AK3-A
AK3-B
AK4
AK5
AEK3
AEK4
AEK3EC
N (%) M (%) C (%)
37 33 30
34 34 32
33 33 34
32 19 49
42 — 58
33 — 67
— — 100
19 — 81
AEK4EC
AK4
AK5
AEK4
AEK4EC
14 — 86
28 — 72
19 26 55
31 — 69
— 25 75
Group B (NCD) CS
+2
+3
Peptides
AK3-A
AK3-B
AK4
AK5
AEK3
AEK4
AEK3EC
N (%) M (%) C (%)
26 45 29
25 43 32
23 28 49
7 17 76
63 — 37
51 — 49
— — 100
located on different sites randomly. However, the NCD in Group B varies depending on the identity of the peptides. In group B, the NCD changes greatly for AKn ions. For the shorter [AK3 + 2H]2+ ion, group B contains a higher percentage of structures whose net charge is located in the middle of the peptide ion (i.e. the charges are located on K3K13, 45% (AK3-A) and 43% (AK3-B)) when compared to the TNCD (33%), since charges that are separated can reduce Coulombic repulsion more efficiently. This further suggests the existence of Coulombic repulsion for short peptide ions with multiple charges. For longer [AKn + 2H]2+ and [AKn + 3H]3+ ions, AK4 and AK5 have a higher percentage of structures with net charge at the C-terminal side in group B compared to the TNCD (Table 1). This is because AK4 and AK5 ions have high helical propensity at the N-terminal side (Fig. 3), and the net charge at the C-terminal side can stabilize the helix macrodipole59,60. Therefore, owing to the preference of net charge site at the C-terminal side and with less side chain – side chain interactions, AK4 and AK5 ions prefer conformations with high helical propensity at the N-terminal side and charge-solvated structure at the C-terminal side. Owing to the restricted conformation space of peptides with high helical propensity, the CCS profiles for AK4 and AK5 ions are narrow and symmetrical (Fig. 1). On the other hand, for [AK3 + 2H]2+ ions, although the middle net
This journal is © The Royal Society of Chemistry 2015
charge site (K3K8) is preferred, N- and C- net charge sites (net charge is located on the N- and C-terminal side) are still competitive, so the possibility of multiple charge sites results in multiple peaks (Fig. 1). Unlike AKn ions, the NCD for AEKn ions depends on charge states. [AEKn + 2H]2+ ions have a higher percentage of conformations whose net charge lies on the N-terminal side in group B compared to TNCD (Table 1). Combined with the charge-solvated structure for [AEKn + 2H]2+ (Fig. 4A), the result further suggests that N-terminal side net charge does not promote formation of helical conformations. Conversely, the net charge site in group B is almost the same as the TNCD for [AEKn + 3H]3+ ions. This suggests that if a peptide does not have high helical propensity, even though its conformation is extended, the C-terminal side is not the preferred position for net charges, viz. the charge sites are random. Meaning of model peptides trend lines In Fig. 1B and 3B, the trend lines for [AKn + 2H]2+ (blue), [AKn + 3H]3+ (blue) and [AEKn + 3H]3+ (red) ions deviate sharply from that of globular conformations, whereas the trend line for [AEKn + 2H]2+ (red) ions deviate only slightly and is parallel to the globular trend line as repeat unit n increases. These trends are different from the trend in helical content with
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
Fig. 5
Analyst
The number of helical residues for AKn (blue triangle) and AEKn ions (red diamond).
repeat unit n (Fig. 1C and 3C). To study the meaning of model peptide trend line, the number of helical residues (residues included in the helical region) was calculated using eqn (2) and plotted in Fig. 5. Note that the number of helical residues for [AKn + 2H]2+ ions (blue triangle) increases linearly with repeat unit n, whereas these values change very little for the [AEKn + 2H]2+ ions (red diamond, n = 3–5) which matches the trend lines for [AKn + 2H]2+ (Fig. 1B) and [AEKn + 2H]2+ (Fig. 3B) ions well. A similar relationship is also observed for [AKn + 3H]3+ and [AEKn + 3H]3+. Therefore, the trend lines for model peptide ions in Fig. 1B and 3B reflect the change in the number of helical residues with repeat unit (molecular weight).
advantage over backbone N and O atoms in terms of proton affinity.62,63 All the backbone and side chain carbonyl groups may be charge carriers. Under this condition, the CCS profiles of AKn and AEKn ions change drastically. For AEKn ions, the peak number increases from a single peak in the unmodified peptide to multiple peaks after acetylation. For AKn ions, although the peak shapes do not change dramatically, the peak width broadens, indicating the presence of multiple conformers. The CCS profiles for the modified peptides clearly show that charge sites and side chain interactions do affect peptide conformation. Differences in charge solvation and Coulombic repulsion, owing to multiple potential sites of protonation, has been shown to have a direct influence on the heterogeneity of peptide ion conformations.
Chemically modified AKn and AEKn ions The simulation results suggest that for both AKn and AEKn ions, charge states, charge sites, and side chain interactions all affect the conformation of peptide ions. Further evidence in support of this general statement can be found in preliminary results for this same series of peptide ions where E and K side chains have been modified, viz. methylation of E and K as well as acetylation of K. Selected CCS profiles for the modified peptides are shown in Fig. S4 (see ESI†). The CCS profiles show that different chemical modifications of charge carrying side chains have different effects on peptide structure, which is mostly triggered by the changes in the side chain interactions and charge sites. Fig. S4† shows that after methylation of lysine (K-(CH3)2) and glutamic acid residues (E-(CH3)n), CCS increases. Methylation of lysine and glutamic acid does not change the charge carrier position, i.e., lysine side chains are still the preferred charge carrier due to their high proton affinity.61 Hence, alteration of the side chain – backbone/side chain interaction may be the origin of the change in peptide structure. After methylation, the added methyl groups on lysine may inhibit the side chain – backbone and side chain – side chain interactions due to steric effects. As a result, the charge solvation effect decreases. A similar effect was also observed in the case of methylated glutamic acid residues. After acetylation of the lysine chains (Fig. S2,† (K-Ac)n and (K-Ac)n & (E-CH3)n), the lysine side chains have no apparent
Analyst
Conclusion Here, we examined the effects of charge states, charge sites and polar side chain interactions on peptide ion structure using two model peptide series, AKn and AEKn. Results obtained from ion mobility and MD simulations can be summarized as follows: (1). Higher charge state leads to increased Coulombic repulsion and if the Coulombic repulsion dominates in the peptide, an extended random coil structure is formed. For example, for [AEKn + 3H]3+ ions, with the support of Coulombic repulsion, cross-linking side chain interactions decrease greatly. As a result, the side chain – side chain interactions help to support an extended random coil. (2). Side chain – side chain interactions at the position (i, i + 5) for K–K and (i, i + 3) for E–K do not promote formation of a helical structure. Side chain – backbone interaction tends to break backbone H-bonds and leads to a charge-solvated structure. (3). The position of charges is significant to peptide structure: a C-terminal half net charge that can stabilize the helix macrodipole results in a helical structure while an N-terminal half net charge results in a random coil structure. Therefore, considering all the factors affecting peptide structure, preferred conformations of AKn ions have higher helical propensity at the N-terminal side and charge-solvated structure at the C-terminal side; [AEKn + 2H]2+ ions prefer to be in charge-
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Analyst
solvated conformations owing to the cross-linking side chain – side chain and side chain – backbone interactions, while [AEKn + 3H]3+ ions prefer to be in extended random coil conformations owing to the support of side chain – side chain interactions under higher Coulombic repulsion. Results from the modification of polar side chains of AKn and AEKn ions further indicate that side chain interaction and charge sites significantly influence peptide structure. Methylation of lysine and glutamic acid side chains may increase helical propensity owing to the steric effect of bulky methyl groups which reduces charge solvation effects. On the other hand, the possibility of multiple charge locations leads to more heterogeneous conformations due to differences in charge solvation and Coulombic repulsion. Lastly, comparison of the helical content of gas-phase ions with different charge state (+1, +2 and +3) to the helical content of solution-phase ions shows that [AKn + 2H]2+ has similar helical content to solution-phase AKn ions while the helical content of [AEKn + 3H]3+is similar to that of solutionphase AEKn ions.
Acknowledgements The authors would like to thank the National Science Foundation (CHE-054587) and the Department of Energy, Basic Energy Sciences (BES DE-FG02-04ER15520) for funding this work.
References 1 G. Chassaing, O. Convert and S. Lavielle, Eur. J. Biochem./ FEBS, 1986, 154, 77–85. 2 J. R. Macdonald and W. C. Johnson, Protein Sci., 2001, 10, 1172–1177. 3 Q. Shao, Y. B. Fan, L. J. Yang and Y. Q. Gao, J. Chem. Phys., 2012, 136. 4 R. Beveridge, Q. Chappuis, C. Macphee and P. Barran, Analyst, 2013, 138, 32–42. 5 K. R. Shoemaker, P. S. Kim, D. N. Brems, S. Marqusee, E. J. York, I. M. Chaiken, J. M. Stewart and R. L. Baldwin, Proc. Natl. Acad. Sci. U. S. A., 1985, 82, 2349–2353. 6 P. Luo and R. L. Baldwin, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 4930–4935. 7 J. M. Scholtz, H. Qian, V. H. Robbins and R. L. Baldwin, Biochemistry, 1993, 32, 9668–9676. 8 J. S. Smith and J. M. Scholtz, Biochemistry, 1998, 37, 33–40. 9 T. Ishimoto, H. Tokiwa, H. Teramae and U. Nagashima, J. Chem. Phys., 2005, 122. 10 M. Karplus and J. A. McCammon, Nat. Struct. Biol., 2002, 9, 646–652. 11 B. T. Ruotolo and D. H. Russell, J. Phys. Chem. B, 2004, 108, 15321–15331.
This journal is © The Royal Society of Chemistry 2015
Paper
12 B. Schuler, E. A. Lipman, P. J. Steinbach, M. Kumke and W. A. Eaton, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 2754– 2759. 13 B. Schuler and W. A. Eaton, Curr. Opin. Struct. Biol., 2008, 18, 16–26. 14 M. Karplus and G. A. Petsko, Nature, 1990, 347, 631–639. 15 S. Hwang, Q. Shao, H. Williams, C. Hilty and Y. Q. Gao, J. Phys. Chem. B, 2011, 115, 6653–6660. 16 O. Glatter and O. Kratky, Small angle x-ray scattering, Academic Press, London, New York, 1982. 17 G. L. Hura, A. L. Menon, M. Hammel, R. P. Rambo, F. L. Poole, S. E. Tsutakawa, F. E. Jenney, S. Classen, K. A. Frankel, R. C. Hopkins, S. J. Yang, J. W. Scott, B. D. Dillard, M. W. W. Adams and J. A. Tainer, Nat. Methods, 2009, 6, 606–U683. 18 L. B. Jensen, K. Mortensen, G. M. Pavan, M. R. Kasimova, D. K. Jensen, V. Gadzhyeva, H. M. Nielsen and C. Foged, Biomacromolecules, 2010, 11, 3571–3577. 19 D. Suckau, Y. Shi, S. C. Beu, M. W. Senko, J. P. Quinn, F. M. Wampler 3rd and F. W. McLafferty, Proc. Natl. Acad. Sci. U. S. A., 1993, 90, 790–793. 20 T. D. Wood, R. A. Chorush, F. M. Wampler 3rd, D. P. Little, P. B. O’Connor and F. W. McLafferty, Proc. Natl. Acad. Sci. U. S. A., 1995, 92, 2451–2454. 21 J. L. Apuy, X. H. Chen, D. H. Russell, T. O. Baldwin and D. P. Giedroc, Biochemistry, 2001, 40, 15164–15175. 22 Y. P. Zhou and R. W. Vachet, Anal. Chem., 2013, 85, 9664– 9670. 23 M. Guidi, U. J. Lorenz, G. Papadopoulos, O. V. Boyarkin and T. R. Rizzo, J. Phys. Chem. A, 2009, 113, 797–799. 24 J. A. Stearns, C. Seaiby, O. V. Boyarkin and T. R. Rizzo, Phys. Chem. Chem. Phys., 2009, 11, 125–132. 25 B. T. Ruotolo and C. V. Robinson, Curr. Opin. Chem. Biol., 2006, 10, 402–408. 26 R. R. Abzalimov, D. A. Kaplan, M. L. Easterling and I. A. Kaltashov, J. Am. Soc. Mass Spectrom., 2009, 20, 1514– 1517. 27 T. Wyttenbach, N. A. Pierson, D. E. Clemmer and M. T. Bowers, Annu. Rev. Phys. Chem., 2014, 65, 175–196. 28 C. S. Creaser, J. R. Griffiths, C. J. Bramwell, S. Noreen, C. A. Hill and C. L. P. Thomas, Analyst, 2004, 129, 984–994. 29 K. J. Lightwahl, B. L. Schwartz and R. D. Smith, J. Am. Chem. Soc., 1994, 116, 5271–5278. 30 J. A. Loo, Int. J. Mass Spectrom., 2000, 200, 175–186. 31 R. D. Smith, Int. J. Mass Spectrom., 2000, 200, 509–544. 32 J. M. Daniel, S. D. Friess, S. Rajagopalan, S. Wendt and R. Zenobi, Int. J. Mass Spectrom., 2002, 216, 1–27. 33 N. A. Pierson, L. Chen, S. J. Valentine, D. H. Russell and D. E. Clemmer, J. Am. Chem. Soc., 2011, 133, 13810–13813. 34 O. S. Skinner, K. Breuker and F. W. McLafferty, J. Am. Soc. Mass Spectrom., 2013, 24, 807–810. 35 J. A. Silveira, K. L. Fort, D. Kim, K. A. Servage, N. A. Pierson, D. E. Clemmer and D. H. Russell, J. Am. Chem. Soc., 2013, 135, 19147–19153. 36 C. Bleiholder, T. Wyttenbach and M. T. Bowers, Int. J. Mass Spectrom., 2011, 308, 1–10.
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
37 S. E. Anderson, C. Bleiholder, E. R. Brocker, P. J. Stang and M. T. Bowers, Int. J. Mass Spectrom., 2012, 330, 78–84. 38 C. Bleiholder, S. Contreras and M. T. Bowers, Int. J. Mass Spectrom., 2013, 354, 275–280. 39 C. Bleiholder, S. Contreras, T. D. Do and M. T. Bowers, Int. J. Mass Spectrom., 2013, 345, 89–96. 40 M. F. Mesleh, J. M. Hunter, A. A. Shvartsburg, G. C. Schatz and M. F. Jarrold, J. Phys. Chem., 1996, 100, 16082– 16086. 41 R. V. de Carvalho, D. Lopez-Ferrer, K. S. Guimaraes and R. D. Lins, J. Comput. Chem., 2013, 34, 1707–1718. 42 A. Chakrabartty and R. L. Baldwin, Adv. Protein Chem., 1995, 46, 141–176. 43 J. M. Scholtz, H. Qian, E. J. York, J. M. Stewart and R. L. Baldwin, Biopolymers, 1991, 31, 1463–1470. 44 C. A. Rohl, J. M. Scholtz, E. J. York, J. M. Stewart and R. L. Baldwin, Biochemistry, 1992, 31, 1263–1269. 45 J. R. McLean, J. A. McLean, Z. X. Wu, C. Becker, L. M. Perez, C. N. Pace, J. M. Scholtz and D. H. Russell, J. Phys. Chem. B, 2010, 114, 809–816. 46 V. Munoz and L. Serrano, Biopolymers, 1997, 41, 495–509. 47 B. T. Ruotolo, J. L. P. Benesch, A. M. Sandercock, S. J. Hyung and C. V. Robinson, Nat. Protoc., 2008, 3, 1139– 1152. 48 S. J. Valentine, A. E. Counterman and D. E. Clemmer, J. Am. Soc. Mass Spectrom., 1999, 10, 1188–1211. 49 D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling, J. Wang, R. E. Duke, R. Luo, R. C. Walker, W. Zhang, K. M. Merz, B. Roberts, B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry, K. F. Wong, F. Paesani, J. Vanicek, J. Liu, X. Wu, S. R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui, D. R. Roe, D. H. Mathews, M. G. Seetin,
Analyst
Analyst
50 51 52
53
54 55 56 57 58 59 60 61 62
63
C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko and P. A. Kollman, AMBER 11, University of California, San Francisco, 2010. V. Hornak, R. Abel, A. Okur, B. Strockbine, A. Roitberg and C. Simmerling, Proteins, 2006, 65, 712–725. W. Kabsch and C. Sander, Biopolymers, 1983, 22, 2577– 2637. R. P. Joosten, T. A. te Beek, E. Krieger, M. L. Hekkelman, R. W. Hooft, R. Schneider, C. Sander and G. Vriend, Nucleic Acids Res., 2011, 39, D411–D419. K. E. Van Holde, W. C. Johnson and P. S. Ho, Principles of Physical Biochemistry, Prentice Hall, Upper Saddle River, NJ, 1998. A. M. Klibanov, Nature, 2001, 409, 241–246. C. Mattos and D. Ringe, Curr. Opin. Struct. Biol., 2001, 11, 761–764. B. T. Ruotolo, G. F. Verbeck, L. M. Thomson, K. J. Gillig and D. H. Russell, J. Am. Chem. Soc., 2002, 124, 4214–4215. C. S. Hoaglund-Hyzer, A. E. Counterman and D. E. Clemmer, Chem. Rev., 1999, 99, 3037–3080. R. Sudha, M. Kohtani, G. A. Breaux and M. F. Jarrold, J. Am. Chem. Soc., 2004, 126, 2777–2784. A. E. Counterman and D. E. Clemmer, J. Am. Chem. Soc., 2001, 123, 1490–1498. R. R. Hudgins, M. A. Ratner and M. F. Jarrold, J. Am. Chem. Soc., 1998, 120, 12974–12975. M. F. Bush, J. Oomens and E. R. Williams, J. Phys. Chem. A, 2009, 113, 431–438. V. Addario, Y. Z. Guo, I. K. Chu, Y. Ling, G. Ruggerio, C. F. Rodriquez, A. C. Hopkinson and K. W. M. Siu, Int. J. Mass Spectrom., 2002, 219, 101–114. J. Evans, G. Nicol and B. Munson, J. Am. Soc. Mass Spectrom., 2000, 11, 789–796.
This journal is © The Royal Society of Chemistry 2015
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
PAPER
Cite this: DOI: 10.1039/c5an00826c
View Journal
Effects of charge states, charge sites and side chain interactions on conformational preferences of a series of model peptide ions† Chunying Xiao, Lisa M. Pérez and David H. Russell* The effects of charge states, charge sites and side chain interactions on conformational preferences of gas-phase peptide ions are examined by ion mobility-mass spectrometry (IM-MS) and molecular dynamics (MD) simulations. Collision cross sections (CCS) of [M + 2H]2+ and [M + 3H]3+ ions for a series of model peptides, viz. Ac-(AAKAA)nY-NH2 (AKn, n = 3–5) and Ac-Y(AEAAKA)nF-NH2 (AEKn, n = 2–5) are measured by using IM-MS and compared with calculated CCS for candidate ions generated by MD simu-
Received 27th April 2015, Accepted 10th June 2015 DOI: 10.1039/c5an00826c www.rsc.org/analyst
lations. The results show that charge states, charge sites and intramolecular charge solvation are important determinants of conformer preference for AKn and AEKn ions. For AKn ions, there is a strong preference for helical conformations near the N-terminus and charge-solvated conformations near the C-terminus. For [AEKn + 2H]2+ ions, conformer preferences appear to be driven by charge solvation, whereas [AEKn + 3H]3+ ions favor more extended coil-type conformations.
Introduction Peptides and proteins are highly dynamic, sampling many conformations on rapid time scales, and both the dynamics and conformational preferences are sensitive to the local environment.1–4 Conformational preferences and the dynamics of interconversions among the different conformations are highly dependent on intramolecular interactions, i.e., van der Waals interactions, hydrogen bonding, and electrostatic interactions. The presence of water as well as cations and anions also plays a key role in determining conformational preferences, both in terms of the hydrophobic effects as well as hydrophilic interactions, viz. solvent-accessibility to hydrophilic side chains of histidine, lysine, arginine, and interactions (salt-bridges) of these charge sites with oppositely charged aspartate and glutamate (R–COO−) as well as asparagine and glutamine.5,6 The effects of intramolecular hydrophilic interactions on the conformational preferences of peptides/proteins have been extensively studied7–9 and advances in experimental techniques and molecular dynamics (MD) simulations are providing new approaches to studies of both thermodynamic stability and folding/unfolding kinetics of native and non-native conformations.8,10 Although MD simulations are widely used for studies of peptide/protein con-
Texas A&M University, Department of Chemistry, College Station, Texas 77843, USA. E-mail: [email protected] † Electronic supplementary information (ESI) available. See DOI: 10.1039/ c5an00826c
This journal is © The Royal Society of Chemistry 2015
formational preferences, simulations only provide candidate conformations that must be evaluated against experimental data, i.e., from circular dichroism (CD),11 fluorescence and fluorescence resonance energy transfer (FRET),12,13 nuclear magnetic resonance (NMR),14,15 X-ray diffraction (XRD),14 and small-angle X-ray scattering (SAXS).16–18 There is increasing awareness that mass spectrometry (MS) approaches add new dimensions for understanding peptide/protein structure and extending our understanding of structure/function relationships, especially for studies of systems that are composed of multiple conformations. MS-based approaches for studies of peptide/protein conformations have evolved to include H/D exchange,19,20 chemical labeling,21,22 IR-UV double resonance spectroscopy23,24 and tandem MS.25,26 IM-MS combined with MD simulation complements the more traditional approaches mentioned above27 since it provides a direct determination of the conformational heterogeneity of the ion population, and it can be adapted to high throughput workflows for screening complex peptide/ protein libraries.25,28 Electrospray ionization (ESI) is clearly the ionization method of choice for structural MS since it has been shown to maintain noncovalent interactions,29–32 and even to retain solution phase secondary structure.33 Recently, Breuker and coworkers studied protein structures in the solution and gas phase, and demonstrated that salt bridges can help to stabilize native/compact structure on short time scales when the proteins are transferred from solution phase to the gas phase.34,35 A significant advantage of ESI is the ability to produce mass spectra dominated by multiply-charged ions;
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
however, it is difficult to determine the location of the charge for molecules that contain multiple possible charge sites. IM-MS is based on the measurement of ion-neutral CCS, which reflects the size and three-dimensional shape of an ion, and provides a direct measurement on population heterogeneity, viz. the number of peaks in the IM arrival-time distributions (ATD). A priori assignment of IM CCSs is a daunting task, viz., currently there exist no guiding principles for correlating the measured CCS for an ion to specific 3-D shapes. In some cases it is possible to correlate CCS with 3-D shapes derived from other experimental measurements, specifically XRD or NMR;36–39 however, MD simulations provide a broader sampling of candidate conformations of the ions whose CCS values can be calculated by using MOBCAL40 (or similar methods).36,41 The “theoretical” CCS obtained using MD simulations and MOBCAL can then be compared to the experimental CCS. This study employs two series of peptides to illustrate the data mining made possible by integration of IM-MS and MD simulation. The peptide sequences are Ac-(AAKAA)nY-NH2 (AKn series, n = 3–5) and Ac-Y(AEAAKA)nF-NH2 (AEKn series, n = 2–5); Ac- indicates acetylation at the N-termini and –NH2 indicates amidation at the C-termini. The N-termini and C-termini of the peptides in each series were protected to reduce the likelihood of collapse of the helix by having charge(s) located at the termini.42 The model peptides used in this study have been used previously by several groups to probe helical propensities of alanine-containing peptides in solution43,44 and in low dielectric gas-phase (solvent-free) environments.45 Previous CD studies show that in solution, the helical content of both AKn and AEKn increases as n increases,43,44,46 and the helical content of AEKn is higher than that of AKn for peptides of comparable length owing to the stabilization afforded by salt bridges. In prior IM-MS studies of the singly-charged ions for the two peptides, it was found that gas-phase ions of AKn (n = 3–6) have higher helical content (ca. 60% helical) than AEKn, and helical propensity is the highest when charge is aligned with the helix macrodipole, i.e., the charge is located on or near the C-termini.45 Despite extensive studies of the AKn and AEKn series, the following questions still remain unanswered: (i) how do charge sites and charge states affect conformer preferences; (ii) how do intramolecular interactions involving the polar side chains affect conformer preferences; and (iii) is there any relationship between solution-phase and gas-phase conformations? Here, ESI-IM-MS in conjunction with MD simulations was used to evaluate the conformational preferences of the two model peptide series in an effort to address these questions.
Experimental section Sample preparation Model peptides AKn (n = 3–5) and AEKn (n = 2–5) were purchased from Shanghai Mocell Biotech Co., Ltd and used without further purification. Solutions of the model peptides
Analyst
Analyst
(2 μM) were prepared using 10 : 90 (v/v) methanol/buffer (buffer: 50 mM ammonium acetate, 25 mM imidazole; pH was adjusted to 7 with 0.1 M acetic acid). ESI-IM-MS All mobility profiles were acquired on a Waters Synapt HDMS G2 mass spectrometer (Manchester, U.K.). Ions were produced by nano-ESI with a source temperature of 120 °C and capillary voltage of 1.5 kV, sampling cone voltage of 40 V and extraction cone voltage of 4 V. For ion mobility experiments, the traveling wave ion mobility cell was maintained at 3.01 mbar N2. The travelling wave velocity is 500 m s−1 and wave height is 20 V for [AKn + 3H]3+; wave velocity and wave height are 300 m s−1 and 20 V respectively for both [AKn + 2H]2+ and [AEKn + 3H]3+; wave velocity and wave height are 800 m s−1 and 40 V respectively for [AEKn + 2H]2+ due to the difference in mobility of different peptides. CCS calibration was performed as previously described by Ruotolo et al.47 Tryptically digested peptides of cytochrome c and myoglobin were used as calibration standards and literature CCS values of doubly-charged peptide ions are cited from the Clemmer group’s CCS database.48 It should be noted that the triply-protonated model peptides are also calibrated using doubly-charged ions because currently there is no standard calibrant database for triply-protonated peptide ions. To probe the stability of the model peptides, solvents conditions were screened to ensure CCS profiles are not influenced by the solvents (data not shown). Instrumental conditions such as capillary voltage, sampling cone voltage and extraction cone voltage, were also screened to ensure arrival time distribution (ATD) did not change under various conditions. Calculation of helical content and number of helical residues Helical content ð% Þ ¼
Ωobs Ωglob Ωhelix Ωglob
100
ð1Þ
Number of helical residues ¼ Helical content total residue number
ð2Þ
Ωobs is the CCS determined by the centroid of the peak profiles. Ωhelix is the calculated CCS for a rigid α-helix cited from literature.45 Ωglob is the calculated globular CCS using the standard globular trend line equations: y ¼ 0:1364x þ 105:99 y ¼ 0:1272x þ 134
For þ 2 species For þ 3 species
ð3Þ ð4Þ
here, x is the mass of peptides, and y is the CCS of globular structure. The standard trend line for doubly-protonated peptide ions (grey in Fig. 1B and 3B, upper) was drawn based on the literature data of doubly-charged ions.48 The standard trend line for triply-protonated ions (grey in Fig. 1B and 3B, lower) was drawn based on standard triply-protonated peptide ions pre-
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Analyst
Paper
Fig. 1 CCS profiles and helical content for AKn. A. CCS profiles for AKn ions. Green peaks are the deconvoluted peak profiles for [AK3 + 2H]2+. “A” and “B” are the higher and lower abundance peaks for [AK3 + 2H]2+. B. CCS vs. molecular weight of AKn ions. Grey points are standard peptide data representing CCS for peptide ions that prefer compact globular conformations in the gas-phase. The doubly-charged standard data is cited from the database of the Clemmer group and triply-charged standard data are from digested proteins. C. Helical content of AKn ions in the solution phase (SP, black) and gas-phase (GP, blue).
pared from the miscleavage of trypsin digestion of proteins (ubiquitin, lysozyme (from chicken egg white), aldolase, enolase, avidin (from egg white), albumin (from chicken egg), ribonuclease (bovin pancrease), myoglobin (horse), cytocrome c, BSA (horse)). All proteins were obtained from Sigma-Aldrich and used without further purification. Molecular dynamics (MD) simulations Candidate conformations of AKn and AEKn ions were generated by simulated annealing (SA) using AMBER1149 with the FF99SB force field.50 For the AKn and AEKn peptide series, both the N-termini and C-termini are protected (acetylated and amidated, respectively), thus the lysine side chains, which have high proton affinities, were considered as the charge carrying sites. In cases where the number of lysines exceeded the numbers of protons, all possible proton positions were considered. The α-helical and fully extended conformations were used as starting structures for SA. During one annealing cycle, the system was heated from 300 K to 1000 K and then cooled down to 300 K over 8400 fs with a time step of 1 fs. After each annealing cycle, the structure energy was minimized. SA was performed for 300 annealing cycles producing 300 minimized structures per simulation. To ensure proper sampling of the
This journal is © The Royal Society of Chemistry 2015
conformational space, a second set of simulations was performed starting with the lowest energy structures from the first set of simulations. For [AEK4 + 2H]2+, two sets of simulations did not produce enough sampling, i.e., the candidate structures did not vary much, therefore, a third set of simulations was performed. The highest energy conformation and a randomly selected middle energy conformation with CCSs in the range of ±2% of the experimental value from the first and second set of simulations were chosen as the starting structures of the third set of simulations. CCSs were calculated by the trajectory method in MOBCAL.40 The secondary structure was calculated using DSSP program.51,52 When processing the simulation data, filtering by CCS and RMSD cluster analysis was performed on the candidate conformations. The procedure for making clusters is described in Fig. S1 (see ESI†). First, all generated conformations that have CCSs within ±2% of the experimental CCS value for the doubly-charged ions form family 1 (in the case of triplycharged ions, ±5% of the experimental data form family 1 because the triply-charged ions were calibrated by doublycharged standard ions which may result in higher error). Structures in family 1 were then clustered based on backbone RMSD (Root-mean-square deviation). Cluster 1 consists of all
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
conformations in family 1 that have an RMSD of the backbone atoms less than a specified value (RMSD cutoff is typically 4–4.5 Å2) when compared to the lowest energy structure in family 1. All of the conformations with an RMSD higher than the specified value make family 2. Cluster 2 contains all structures with an RMSD of the backbone atoms below the cutoff when compared to the lowest energy structure in family 2 and all the conformations with an RMSD higher than the cutoff make family 3. The procedure was repeated for the remaining structures until all the structures were allocated into a cluster. Therefore, the reference structure for cluster 1 has the lowest energy, the reference structure for cluster 2 has higher energy than that for cluster 1 and the reference structure for the last cluster has the highest energy. It should be noted that due to the different size of peptides, the RMSD cutoff was adjusted to produce a reasonable number of clusters ( m/2 denotes that the net charge is located on the C-terminal half (C)
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Theoretical net charge distribution (TNCD) CS
+2
Peptides N (%) M (%) C (%)
+3 AK3
AK4
AK5
AEK3
AEK4
AEK3EC
AEK4EC
AK4
AK5
AEK4
AEK4EC
33 33 33
33 33 33
40 20 40
33 — 67
33 — 67
— — 100
17 — 83
50 — 50
40 20 40
25 0 75
0 25 75
AEK4EC
AK4
AK5
AEK4
AEK4EC
57 — 43
46 20 34
25 — 75
— 25 75
Group A (NCD) CS
+2
+3
Peptides
AK3-A
AK3-B
AK4
AK5
AEK3
AEK4
AEK3EC
N (%) M (%) C (%)
37 33 30
34 34 32
33 33 34
32 19 49
42 — 58
33 — 67
— — 100
19 — 81
AEK4EC
AK4
AK5
AEK4
AEK4EC
14 — 86
28 — 72
19 26 55
31 — 69
— 25 75
Group B (NCD) CS
+2
+3
Peptides
AK3-A
AK3-B
AK4
AK5
AEK3
AEK4
AEK3EC
N (%) M (%) C (%)
26 45 29
25 43 32
23 28 49
7 17 76
63 — 37
51 — 49
— — 100
located on different sites randomly. However, the NCD in Group B varies depending on the identity of the peptides. In group B, the NCD changes greatly for AKn ions. For the shorter [AK3 + 2H]2+ ion, group B contains a higher percentage of structures whose net charge is located in the middle of the peptide ion (i.e. the charges are located on K3K13, 45% (AK3-A) and 43% (AK3-B)) when compared to the TNCD (33%), since charges that are separated can reduce Coulombic repulsion more efficiently. This further suggests the existence of Coulombic repulsion for short peptide ions with multiple charges. For longer [AKn + 2H]2+ and [AKn + 3H]3+ ions, AK4 and AK5 have a higher percentage of structures with net charge at the C-terminal side in group B compared to the TNCD (Table 1). This is because AK4 and AK5 ions have high helical propensity at the N-terminal side (Fig. 3), and the net charge at the C-terminal side can stabilize the helix macrodipole59,60. Therefore, owing to the preference of net charge site at the C-terminal side and with less side chain – side chain interactions, AK4 and AK5 ions prefer conformations with high helical propensity at the N-terminal side and charge-solvated structure at the C-terminal side. Owing to the restricted conformation space of peptides with high helical propensity, the CCS profiles for AK4 and AK5 ions are narrow and symmetrical (Fig. 1). On the other hand, for [AK3 + 2H]2+ ions, although the middle net
This journal is © The Royal Society of Chemistry 2015
charge site (K3K8) is preferred, N- and C- net charge sites (net charge is located on the N- and C-terminal side) are still competitive, so the possibility of multiple charge sites results in multiple peaks (Fig. 1). Unlike AKn ions, the NCD for AEKn ions depends on charge states. [AEKn + 2H]2+ ions have a higher percentage of conformations whose net charge lies on the N-terminal side in group B compared to TNCD (Table 1). Combined with the charge-solvated structure for [AEKn + 2H]2+ (Fig. 4A), the result further suggests that N-terminal side net charge does not promote formation of helical conformations. Conversely, the net charge site in group B is almost the same as the TNCD for [AEKn + 3H]3+ ions. This suggests that if a peptide does not have high helical propensity, even though its conformation is extended, the C-terminal side is not the preferred position for net charges, viz. the charge sites are random. Meaning of model peptides trend lines In Fig. 1B and 3B, the trend lines for [AKn + 2H]2+ (blue), [AKn + 3H]3+ (blue) and [AEKn + 3H]3+ (red) ions deviate sharply from that of globular conformations, whereas the trend line for [AEKn + 2H]2+ (red) ions deviate only slightly and is parallel to the globular trend line as repeat unit n increases. These trends are different from the trend in helical content with
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
Fig. 5
Analyst
The number of helical residues for AKn (blue triangle) and AEKn ions (red diamond).
repeat unit n (Fig. 1C and 3C). To study the meaning of model peptide trend line, the number of helical residues (residues included in the helical region) was calculated using eqn (2) and plotted in Fig. 5. Note that the number of helical residues for [AKn + 2H]2+ ions (blue triangle) increases linearly with repeat unit n, whereas these values change very little for the [AEKn + 2H]2+ ions (red diamond, n = 3–5) which matches the trend lines for [AKn + 2H]2+ (Fig. 1B) and [AEKn + 2H]2+ (Fig. 3B) ions well. A similar relationship is also observed for [AKn + 3H]3+ and [AEKn + 3H]3+. Therefore, the trend lines for model peptide ions in Fig. 1B and 3B reflect the change in the number of helical residues with repeat unit (molecular weight).
advantage over backbone N and O atoms in terms of proton affinity.62,63 All the backbone and side chain carbonyl groups may be charge carriers. Under this condition, the CCS profiles of AKn and AEKn ions change drastically. For AEKn ions, the peak number increases from a single peak in the unmodified peptide to multiple peaks after acetylation. For AKn ions, although the peak shapes do not change dramatically, the peak width broadens, indicating the presence of multiple conformers. The CCS profiles for the modified peptides clearly show that charge sites and side chain interactions do affect peptide conformation. Differences in charge solvation and Coulombic repulsion, owing to multiple potential sites of protonation, has been shown to have a direct influence on the heterogeneity of peptide ion conformations.
Chemically modified AKn and AEKn ions The simulation results suggest that for both AKn and AEKn ions, charge states, charge sites, and side chain interactions all affect the conformation of peptide ions. Further evidence in support of this general statement can be found in preliminary results for this same series of peptide ions where E and K side chains have been modified, viz. methylation of E and K as well as acetylation of K. Selected CCS profiles for the modified peptides are shown in Fig. S4 (see ESI†). The CCS profiles show that different chemical modifications of charge carrying side chains have different effects on peptide structure, which is mostly triggered by the changes in the side chain interactions and charge sites. Fig. S4† shows that after methylation of lysine (K-(CH3)2) and glutamic acid residues (E-(CH3)n), CCS increases. Methylation of lysine and glutamic acid does not change the charge carrier position, i.e., lysine side chains are still the preferred charge carrier due to their high proton affinity.61 Hence, alteration of the side chain – backbone/side chain interaction may be the origin of the change in peptide structure. After methylation, the added methyl groups on lysine may inhibit the side chain – backbone and side chain – side chain interactions due to steric effects. As a result, the charge solvation effect decreases. A similar effect was also observed in the case of methylated glutamic acid residues. After acetylation of the lysine chains (Fig. S2,† (K-Ac)n and (K-Ac)n & (E-CH3)n), the lysine side chains have no apparent
Analyst
Conclusion Here, we examined the effects of charge states, charge sites and polar side chain interactions on peptide ion structure using two model peptide series, AKn and AEKn. Results obtained from ion mobility and MD simulations can be summarized as follows: (1). Higher charge state leads to increased Coulombic repulsion and if the Coulombic repulsion dominates in the peptide, an extended random coil structure is formed. For example, for [AEKn + 3H]3+ ions, with the support of Coulombic repulsion, cross-linking side chain interactions decrease greatly. As a result, the side chain – side chain interactions help to support an extended random coil. (2). Side chain – side chain interactions at the position (i, i + 5) for K–K and (i, i + 3) for E–K do not promote formation of a helical structure. Side chain – backbone interaction tends to break backbone H-bonds and leads to a charge-solvated structure. (3). The position of charges is significant to peptide structure: a C-terminal half net charge that can stabilize the helix macrodipole results in a helical structure while an N-terminal half net charge results in a random coil structure. Therefore, considering all the factors affecting peptide structure, preferred conformations of AKn ions have higher helical propensity at the N-terminal side and charge-solvated structure at the C-terminal side; [AEKn + 2H]2+ ions prefer to be in charge-
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Analyst
solvated conformations owing to the cross-linking side chain – side chain and side chain – backbone interactions, while [AEKn + 3H]3+ ions prefer to be in extended random coil conformations owing to the support of side chain – side chain interactions under higher Coulombic repulsion. Results from the modification of polar side chains of AKn and AEKn ions further indicate that side chain interaction and charge sites significantly influence peptide structure. Methylation of lysine and glutamic acid side chains may increase helical propensity owing to the steric effect of bulky methyl groups which reduces charge solvation effects. On the other hand, the possibility of multiple charge locations leads to more heterogeneous conformations due to differences in charge solvation and Coulombic repulsion. Lastly, comparison of the helical content of gas-phase ions with different charge state (+1, +2 and +3) to the helical content of solution-phase ions shows that [AKn + 2H]2+ has similar helical content to solution-phase AKn ions while the helical content of [AEKn + 3H]3+is similar to that of solutionphase AEKn ions.
Acknowledgements The authors would like to thank the National Science Foundation (CHE-054587) and the Department of Energy, Basic Energy Sciences (BES DE-FG02-04ER15520) for funding this work.
References 1 G. Chassaing, O. Convert and S. Lavielle, Eur. J. Biochem./ FEBS, 1986, 154, 77–85. 2 J. R. Macdonald and W. C. Johnson, Protein Sci., 2001, 10, 1172–1177. 3 Q. Shao, Y. B. Fan, L. J. Yang and Y. Q. Gao, J. Chem. Phys., 2012, 136. 4 R. Beveridge, Q. Chappuis, C. Macphee and P. Barran, Analyst, 2013, 138, 32–42. 5 K. R. Shoemaker, P. S. Kim, D. N. Brems, S. Marqusee, E. J. York, I. M. Chaiken, J. M. Stewart and R. L. Baldwin, Proc. Natl. Acad. Sci. U. S. A., 1985, 82, 2349–2353. 6 P. Luo and R. L. Baldwin, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 4930–4935. 7 J. M. Scholtz, H. Qian, V. H. Robbins and R. L. Baldwin, Biochemistry, 1993, 32, 9668–9676. 8 J. S. Smith and J. M. Scholtz, Biochemistry, 1998, 37, 33–40. 9 T. Ishimoto, H. Tokiwa, H. Teramae and U. Nagashima, J. Chem. Phys., 2005, 122. 10 M. Karplus and J. A. McCammon, Nat. Struct. Biol., 2002, 9, 646–652. 11 B. T. Ruotolo and D. H. Russell, J. Phys. Chem. B, 2004, 108, 15321–15331.
This journal is © The Royal Society of Chemistry 2015
Paper
12 B. Schuler, E. A. Lipman, P. J. Steinbach, M. Kumke and W. A. Eaton, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 2754– 2759. 13 B. Schuler and W. A. Eaton, Curr. Opin. Struct. Biol., 2008, 18, 16–26. 14 M. Karplus and G. A. Petsko, Nature, 1990, 347, 631–639. 15 S. Hwang, Q. Shao, H. Williams, C. Hilty and Y. Q. Gao, J. Phys. Chem. B, 2011, 115, 6653–6660. 16 O. Glatter and O. Kratky, Small angle x-ray scattering, Academic Press, London, New York, 1982. 17 G. L. Hura, A. L. Menon, M. Hammel, R. P. Rambo, F. L. Poole, S. E. Tsutakawa, F. E. Jenney, S. Classen, K. A. Frankel, R. C. Hopkins, S. J. Yang, J. W. Scott, B. D. Dillard, M. W. W. Adams and J. A. Tainer, Nat. Methods, 2009, 6, 606–U683. 18 L. B. Jensen, K. Mortensen, G. M. Pavan, M. R. Kasimova, D. K. Jensen, V. Gadzhyeva, H. M. Nielsen and C. Foged, Biomacromolecules, 2010, 11, 3571–3577. 19 D. Suckau, Y. Shi, S. C. Beu, M. W. Senko, J. P. Quinn, F. M. Wampler 3rd and F. W. McLafferty, Proc. Natl. Acad. Sci. U. S. A., 1993, 90, 790–793. 20 T. D. Wood, R. A. Chorush, F. M. Wampler 3rd, D. P. Little, P. B. O’Connor and F. W. McLafferty, Proc. Natl. Acad. Sci. U. S. A., 1995, 92, 2451–2454. 21 J. L. Apuy, X. H. Chen, D. H. Russell, T. O. Baldwin and D. P. Giedroc, Biochemistry, 2001, 40, 15164–15175. 22 Y. P. Zhou and R. W. Vachet, Anal. Chem., 2013, 85, 9664– 9670. 23 M. Guidi, U. J. Lorenz, G. Papadopoulos, O. V. Boyarkin and T. R. Rizzo, J. Phys. Chem. A, 2009, 113, 797–799. 24 J. A. Stearns, C. Seaiby, O. V. Boyarkin and T. R. Rizzo, Phys. Chem. Chem. Phys., 2009, 11, 125–132. 25 B. T. Ruotolo and C. V. Robinson, Curr. Opin. Chem. Biol., 2006, 10, 402–408. 26 R. R. Abzalimov, D. A. Kaplan, M. L. Easterling and I. A. Kaltashov, J. Am. Soc. Mass Spectrom., 2009, 20, 1514– 1517. 27 T. Wyttenbach, N. A. Pierson, D. E. Clemmer and M. T. Bowers, Annu. Rev. Phys. Chem., 2014, 65, 175–196. 28 C. S. Creaser, J. R. Griffiths, C. J. Bramwell, S. Noreen, C. A. Hill and C. L. P. Thomas, Analyst, 2004, 129, 984–994. 29 K. J. Lightwahl, B. L. Schwartz and R. D. Smith, J. Am. Chem. Soc., 1994, 116, 5271–5278. 30 J. A. Loo, Int. J. Mass Spectrom., 2000, 200, 175–186. 31 R. D. Smith, Int. J. Mass Spectrom., 2000, 200, 509–544. 32 J. M. Daniel, S. D. Friess, S. Rajagopalan, S. Wendt and R. Zenobi, Int. J. Mass Spectrom., 2002, 216, 1–27. 33 N. A. Pierson, L. Chen, S. J. Valentine, D. H. Russell and D. E. Clemmer, J. Am. Chem. Soc., 2011, 133, 13810–13813. 34 O. S. Skinner, K. Breuker and F. W. McLafferty, J. Am. Soc. Mass Spectrom., 2013, 24, 807–810. 35 J. A. Silveira, K. L. Fort, D. Kim, K. A. Servage, N. A. Pierson, D. E. Clemmer and D. H. Russell, J. Am. Chem. Soc., 2013, 135, 19147–19153. 36 C. Bleiholder, T. Wyttenbach and M. T. Bowers, Int. J. Mass Spectrom., 2011, 308, 1–10.
Analyst
View Article Online
Published on 10 June 2015. Downloaded by UNIVERSITY OF OTAGO on 31/07/2015 04:25:38.
Paper
37 S. E. Anderson, C. Bleiholder, E. R. Brocker, P. J. Stang and M. T. Bowers, Int. J. Mass Spectrom., 2012, 330, 78–84. 38 C. Bleiholder, S. Contreras and M. T. Bowers, Int. J. Mass Spectrom., 2013, 354, 275–280. 39 C. Bleiholder, S. Contreras, T. D. Do and M. T. Bowers, Int. J. Mass Spectrom., 2013, 345, 89–96. 40 M. F. Mesleh, J. M. Hunter, A. A. Shvartsburg, G. C. Schatz and M. F. Jarrold, J. Phys. Chem., 1996, 100, 16082– 16086. 41 R. V. de Carvalho, D. Lopez-Ferrer, K. S. Guimaraes and R. D. Lins, J. Comput. Chem., 2013, 34, 1707–1718. 42 A. Chakrabartty and R. L. Baldwin, Adv. Protein Chem., 1995, 46, 141–176. 43 J. M. Scholtz, H. Qian, E. J. York, J. M. Stewart and R. L. Baldwin, Biopolymers, 1991, 31, 1463–1470. 44 C. A. Rohl, J. M. Scholtz, E. J. York, J. M. Stewart and R. L. Baldwin, Biochemistry, 1992, 31, 1263–1269. 45 J. R. McLean, J. A. McLean, Z. X. Wu, C. Becker, L. M. Perez, C. N. Pace, J. M. Scholtz and D. H. Russell, J. Phys. Chem. B, 2010, 114, 809–816. 46 V. Munoz and L. Serrano, Biopolymers, 1997, 41, 495–509. 47 B. T. Ruotolo, J. L. P. Benesch, A. M. Sandercock, S. J. Hyung and C. V. Robinson, Nat. Protoc., 2008, 3, 1139– 1152. 48 S. J. Valentine, A. E. Counterman and D. E. Clemmer, J. Am. Soc. Mass Spectrom., 1999, 10, 1188–1211. 49 D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling, J. Wang, R. E. Duke, R. Luo, R. C. Walker, W. Zhang, K. M. Merz, B. Roberts, B. Wang, S. Hayik, A. Roitberg, G. Seabra, I. Kolossváry, K. F. Wong, F. Paesani, J. Vanicek, J. Liu, X. Wu, S. R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui, D. R. Roe, D. H. Mathews, M. G. Seetin,
Analyst
Analyst
50 51 52
53
54 55 56 57 58 59 60 61 62
63
C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko and P. A. Kollman, AMBER 11, University of California, San Francisco, 2010. V. Hornak, R. Abel, A. Okur, B. Strockbine, A. Roitberg and C. Simmerling, Proteins, 2006, 65, 712–725. W. Kabsch and C. Sander, Biopolymers, 1983, 22, 2577– 2637. R. P. Joosten, T. A. te Beek, E. Krieger, M. L. Hekkelman, R. W. Hooft, R. Schneider, C. Sander and G. Vriend, Nucleic Acids Res., 2011, 39, D411–D419. K. E. Van Holde, W. C. Johnson and P. S. Ho, Principles of Physical Biochemistry, Prentice Hall, Upper Saddle River, NJ, 1998. A. M. Klibanov, Nature, 2001, 409, 241–246. C. Mattos and D. Ringe, Curr. Opin. Struct. Biol., 2001, 11, 761–764. B. T. Ruotolo, G. F. Verbeck, L. M. Thomson, K. J. Gillig and D. H. Russell, J. Am. Chem. Soc., 2002, 124, 4214–4215. C. S. Hoaglund-Hyzer, A. E. Counterman and D. E. Clemmer, Chem. Rev., 1999, 99, 3037–3080. R. Sudha, M. Kohtani, G. A. Breaux and M. F. Jarrold, J. Am. Chem. Soc., 2004, 126, 2777–2784. A. E. Counterman and D. E. Clemmer, J. Am. Chem. Soc., 2001, 123, 1490–1498. R. R. Hudgins, M. A. Ratner and M. F. Jarrold, J. Am. Chem. Soc., 1998, 120, 12974–12975. M. F. Bush, J. Oomens and E. R. Williams, J. Phys. Chem. A, 2009, 113, 431–438. V. Addario, Y. Z. Guo, I. K. Chu, Y. Ling, G. Ruggerio, C. F. Rodriquez, A. C. Hopkinson and K. W. M. Siu, Int. J. Mass Spectrom., 2002, 219, 101–114. J. Evans, G. Nicol and B. Munson, J. Am. Soc. Mass Spectrom., 2000, 11, 789–796.
This journal is © The Royal Society of Chemistry 2015
